AI-Driven Drug Discovery: Pancreatic Cancer Resistance via ENT1 Differential Conformation Tracking

Objetivo

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal cancers, with a five‑year survival rate below 5%, making it a significant global health challenge. Despite extensive research, gemcitabine remains the standard first‑line treatment, but its effectiveness is constrained by rapid resistance and poor patient stratification. The human equilibrative nucleoside transporter 1 (hENT1) plays a key role in gemcitabine uptake and serves as a promising biomarker for predicting treatment response. Recent ligand-bound crystal structures of hENT1 enable atomic‑level investigation of its dynamics and resistance‑linked mutations. The present project aims to uncover the molecular basis of gemcitabine transport and resistance using a unified approach that combines computational structural biology with AI-guided drug design. We will first identify and model hENT1 mutations associated with PDAC progression, then characterize gemcitabine binding to both wild‑type and mutant transporters via molecular docking complemented by extended molecular dynamics simulations. Finally, generative adversarial networks (GANs) will be employed to design gemcitabine analogs with enhanced efficacy against resistant variants. These efforts will elucidate hENT1‑mediated drug transport mechanisms, guide the rational development of next‑generation nucleoside analogs, and support biomarker‑driven therapeutic strategies in PDAC.